Coordinated Coil Scheduling Method of Multiple Production Lines for Cold Rolling Area in Steel Plant

1. A coordinated coil scheduling method of multiple production lines in a cold rolling area in a steel plant, the production lines including an acid rolling line, a continuous annealing line, a hot galvanizing line, and a hot aluminizing and galvanizing line, the method comprising: Step 1: acquiring information about candidate coils to be scheduled for each of multiple production lines in the cold rolling area; Step 2: building a coordinated coil scheduling model of multiple production lines in the cold rolling area in the steel plant; Step 2-1: according to production technological requirements of the production lines, calculating abrasion penalties and changeover penalties incurred by equipment adjustment due to different production requirements, comprising: abrasion penalties of equipment incurred by equipment adjustment due to different widths of two adjacent coils in a production line, abrasion penalties of equipment incurred by equipment adjustment due to different thicknesses of two adjacent coils in a production line, abrasion penalties of equipment incurred by equipment adjustment due to different surface requirements of two adjacent coils in a production line, abrasion penalties of equipment incurred by equipment adjustment due to different annealing temperatures of two adjacent coils in a production line, abrasion penalties of equipment incurred by equipment adjustment due to different weights of the surface plates of two coils in a production line, and abrasion penalties of equipment incurred by equipment adjustment due to different post-processing requirements of two coils in a production line; Step 2-2: among coil schedules corresponding to the production lines in a cold rolling supply chain composed by multiple production lines, an objective function of coordinated coil scheduling of multiple lines in the cold rolling area is built as minimizing the abrasion penalties and changeover penalties incurred by equipment adjustment due to different production requirements, i . e . , minimize ⁢ ⁢ ∑ k ∈ M ⁢ ⁢ ∑ j ∈ N k ⁢ ⁢ ∑ i ∈ N k ⁢ ⁢ F ijk ⁢ x ijk ; where a decision variable of the objective function is represented as x ijk , which indicates whether a coil i is processed immediately before a coil j, in a kth production line, if yes, the decision variable x ijk is assigned as 1, and if not, it is assigned as 0; Step 2-3: constraints comprise ones for each of the multiple lines and cohesion constraints among the production lines; where the constraints for each of the multiple lines include: (1) for each of all actual coils, there must be exactly one coil being processed immediately before it; (2) for each of all the actual coils, there must be exactly one coil being processed immediately after it; (3) changeover in width, thickness, surface requirement, and annealing temperature of lines should not exceed predetermined upper limits; (4) it should be guaranteed that determination to a processing time of coils in each line will not incur idle time between any two adjacent processed coils; (5) for each of the coils, it should be processed exactly once in a production line along its required processing path; where the cohesion constraints among the production lines imply a cohesion relationship between the production lines that the coil cannot be processed in a downstream production line until it is finished in an upstream production line; Step 3: an initial coordinated coil schedule of multiple production lines in the cold rolling area is obtained by using an heuristic algorithm, which includes an initial coil schedule in acid rolling line, an initial coil schedule in continuous annealing line, an initial coil schedule in hot galvanizing line, and an initial schedule of coils in hot aluminizing and galvanizing line; Step 3-1: Determine the initial coil schedule in acid rolling line: an initial coil schedule in the acid rolling line is obtained by sequencing the coils to be processed in a descending order of a surface requirement code, and then sequencing the coils with a same surface requirement in a descending order of their widths, and then sequencing the coils with the same surface requirement and a same width in a descending order of their post-processing codes, and then sequencing the coils with the same surface requirement and the same width and a same post-processing code in an descending order of their annealing temperatures; Step 3-2: calculating a starting time and a finishing time of each coil in the initial coil schedule in the acid rolling line; Step 3-3: determine an initial coil schedule in a continuous annealing line: an initial annealing schedule of coils in a material inventory of continuous annealing line is obtained by sequencing the coils to be annealed in a descending order of their widths, and then sequencing the coils with the same width in a descending order of their annealing temperatures, and then sequencing the coils with the same width and a same annealing temperature in a descending order of their thicknesses, a coil schedule in continuous annealing line is obtained by selecting the coils to be annealed from the coil schedule in the acid rolling line, and sequentially inserting them after the initial annealing schedule of coils in the material inventory of continuous annealing line; Step 3-4: determine an initial coil schedule in the hot galvanizing line: an initial schedule of the coils in the material inventory of hot galvanizing line is obtained, by sequencing the coils in the material inventory of hot galvanizing line in a descending order of their post-processing codes; and then sequencing the coils with the same post-processing code in a descending order of their widths; and then sequencing the coils with the same width and the same post-processing code in a descending order of their thicknesses; and then sequencing the coils with the same post-processing code, the same width and the same thickness in a descending order of their annealing temperatures; an initial coil schedule in the hot galvanizing line is obtained by selecting the coils to be processed in hot galvanizing line from the initial coil schedule in acid rolling line, sequentially inserting them after the initial schedule of the coils in the material inventory of hot galvanizing line; Step 3-5: determine an initial schedule of coils in the hot galvanizing and aluminizing line: an initial schedule of coils in material inventory of the hot galvanizing and aluminizing line is obtained, by sequencing the coils in material inventory of the hot galvanizing and aluminizing line in a descending order of their post-processing codes; and then sequencing the coils with the same post-processing code in a descending order of their widths; and then sequencing the coils with the same post-processing code and the same width in a descending order of their thicknesses; and then sequencing the coils with the same post-processing code, the same width and the same thickness in a descending order of their annealing temperatures; an initial coil schedule in the hot galvanizing and aluminizing line is obtained, by selecting the coils to be processed in hot galvanizing and aluminizing line from the initial coil schedule in acid rolling line, sequentially inserting them after the initial schedule of coils in material inventory of the hot galvanizing and aluminizing line; Step 3-6: calculating the starting time of each coil in the initial coil schedule in the continuous annealing line, the starting time of each coil in the initial coil schedule in the hot galvanizing line, and the starting time of each coil in the initial coil schedule in the hot aluminizing and galvanizing line; Step 3-7: recording an initial coordinated coil schedule of multiple production lines in the cold rolling area, and calculating the objective function value corresponding to the initial coordinated coil schedule of multiple lines in the cold rolling area; Step 4: improving in real time the initial coordinated coil schedule of multiple lines in the cold rolling area; Step 5: sending the improved coordinated coil schedule of multiple lines in the cold rolling area to an automatic control systems (ACS) of each production lines in the cold rolling area, and realizing the coordinated coil scheduling of multiple lines in the cold rolling area, wherein realizing the coordinated coil scheduling includes: processing, by an ACS of the acid rolling line, the coils whose processing route pass the acid rolling line using the acid rolling line according to the coil schedule in acid rolling line which is a part of the improved coordinated coil schedule, processing, by an ACS of the continuous annealing line, the coils whose processing route pass the continuous annealing line using the continuous annealing line according to the coil schedule in continuous annealing line which is a part of the improved coordinated coil schedule, processing, by an ACS of the hot galvanizing line, the coils whose processing route pass the hot galvanizing line using the hot galvanizing line according to the coil schedule in hot galvanizing line which is a part of the improved coordinated coil schedule, and processing, by an ACS of the hot aluminizing and galvanizing line, the coils whose processing route pass the hot aluminizing and galvanizing line using the hot aluminizing and galvanizing line according to the coil schedule in hot aluminizing and galvanizing line which is a part of the improved coordinated coil schedule, and wherein the coordinated coil scheduling improves the quality of the product by having reduced equipment adjustment and reduced abrasion penalties.

2. The method of claim 1 , where the step 4, which improves in real time the initial coordinated coil schedule of multiple lines in the cold rolling area, comprises: Step 4-1: initializing population of the coordinated coil schedule of multiple lines in the cold rolling area, and setting a given maximum iteration number and a given maximum iteration number without improvement, population of the coordinated coil schedule of multiple lines in the cold rolling area comprises the initial coordinated coil schedule of multiple lines in the cold rolling area, and other coil schedules in population of a coordinated coil schedule of multiple lines in the cold rolling area are generated randomly; and during the improvement, an adaptive value of each coordinated coil schedule is calculated according to a way for calculating an objective function of each coordinated coil schedule; Step 4-2: judging whether the given maximum iteration number is reached, and if it is reached, then taking a schedule with a best adaptive value in current population as an optimal schedule, that is, the improved coordinated coil schedule of multiple lines in the cold rolling area, and if not, go to step 4-3; Step 4-3: judging whether the given maximum iteration number without improvement is reached, and if it is reached, then taking the schedule with the best adaptive value in the current population as the optimal schedule, that is, the improved coordinated coil schedule of multiple lines in the cold rolling area, and if not, go to step 4-4; Step 4-4: if the individual with the best adaptive value in the population is not updated, the maximum iteration number without improvement is added by 1; Step 4-5: performing a mutation operation to the population; Step 4-6: performing a crossover operation to the population; Step 4-7: performing a selecting operation on the son generation in the population; Step 4-8: calculating the adaptive value for each individual in the population, that is, calculating the objective function value of coordinated coil schedule of multiple lines in the cold rolling area corresponding to the current population; Step 4-9: performing an increment operation to the current iteration number and updating the population, and then returning to step 4-2.

3. The method of claim 2 , where: the step 4-5, which performs a mutation operation to the population, comprises: V i t = X a t - 1 ⊕ F ⊗ ( X b t - 1 - X c t - 1 ) Δ i t = F ⊗ ( X b t - 1 - X c t - 1 ) ⇔ δ ij t = { x bj t - 1 - x cj t - 1 if ⁢ ⁢ rand ( ) < F 0 otherwise ⁢ ⁢ Δ i t = ( δ i , 0 t , δ i , 1 t , … ⁢ , δ i , n t ) ⁢ ⁢ V i t = X a t - 1 ⊕ Δ i t ⇔ v ij t = x aj t - 1 ⊕ δ ij t = mod ⁡ ( ( x aj t - 1 + δ ij t + n ) , n ) where V i t is an ith variant chromosome in an tth generation, and v ij t is a jth individual in V i t , and X a t−1 , X b t−1 and X c t−1 are three different coordinated coil schedules of multiple lines in the cold rolling area in the population of a t−1th generation, x ij t−1 is a jth individual in X i t −1 , F is a scaling factor, Δ i t is the ith intermediate variant chromosome in the mutation process of the tth generation, δ ij t is a jth individual in the intermediate variant chromosome Δ i t in the mutation process of the tth generation, and mod means modulus.

4. The method of claim 2 , where: the step 4-6, which performs the crossover operation to the population, comprises: generating a trial chromosome U i t , i=1, 2, . . . , PS, and u ij t is an individual in the chromosome U i t ; u ij t = { v ij t if ⁢ ⁢ rand ( ) < CR ⁢ ⁢ or ⁢ ⁢ j = D j ⁢ x ij t - 1 otherwise where CR is in [0,1] which is a crossover parameter to ensure a variety of population, rand( ) generates a group of uniformly distributed random numbers between 0 and 1, D j is a number randomly selected from the set{1, 2, . . . , n}, so as to guarantee that at least one dimension of U i t is different from the ith coordinated coil schedule of multiple lines in the cold rolling area in the t−1th generation X i t−1 ; CR is a self-updating number as the iteration number increases, and is updated as follows: CR = CR min + t * ( CR max - CR min ) T where CR min is a minimum crossover probability, CR max is a maximum crossover probability, t is a current iteration number, and T is a maximum iteration number.